Digital Transformation in National Defense: A Bibliometric Review of Distributed Image Analysis Technology

Abstrak

As remote sensing technology continues to advancing, image analysis systems must efficiently manage substantial volumes of data. The incorporation of deep learning into distributed computing systems has become essential for addressing this demand, facilitating important tasks like object detection and real-time classification. This research methodically outlines the landscape of distributed machine learning for image analysis to pinpoint technological trends that can enhance network-centric imagery intelligence (IMINT). By employing a PRISMA-guided literature review and a bibliometric analysis of 59 publications from Scopus (2015–2026), we chart the field's progression from fundamental distributed computing to optimized, real-time analytical systems. Analyzing keyword co-occurrence and thematic synthesis uncovers significant clusters, including distributed deep learning, privacy-preserving collaboration, and edge-based inference. Based on this examination, we identify and assess key adaptable technologies like split computing, federated learning, and lightweight model optimization for their applicability in operational contexts defined by limited connectivity and decentralized resources. By aligning global research trends with specific national requirements, this review provides a strategic framework for developing resilient, efficient, and adaptive IMINT capabilities within a network-centric paradigm.

Identifier

• DOI: https://doi.org/10.63922/ijevss.v5i02.3170
• OpenAlex: https://openalex.org/W7155385743